1. Field of the Invention
The present invention relates to a method for processing an image, by which image data read by an image reading apparatus such as an image scanner, a copying machine, or a multifunction peripheral is corrected.
2. Description of the Related Art
Currently, when a document that has been bound by saddle stitching, such as a book or a magazine, is read or copied, the document is opened and disposed on a transparent document table of an image scanner, a copying machine, or the like such that a surface to be read or copied faces downward and an image is read from below the document table. A method for processing the image at this time has been disclosed in Japanese Patent Laid-Open No. 2008-54289 and Japanese Patent Laid-Open No. 2003-69824, by which a shadow generated in a bound portion around a boundary between left and right pages of the document, which has been opened and disposed on the document table, is subjected to luminance correction through image processing.
In Japanese Patent Laid-Open No. 2008-54289, however, luminance correction value tables are obtained at two positions that are distant from each other in the vertical direction of the bound portion and a luminance correction value table at an arbitrary position is calculated from the two tables through interpolation calculation. That is, luminance correction values at all pixel positions need to be calculated for each line of an image to be corrected. Therefore, while the quality of an image after the correction is high, a circuit is large in terms of installation of the circuit in a copying machine or a multifunction peripheral (hereinafter referred to as the MFP) in order to perform correction of a shadow during a copying operation, thereby increasing the cost.
In addition, in the example of the related art disclosed in Japanese Patent Laid-Open No. 2003-69824, an appropriate method for determining the size and the aspect ratio of a plurality of blocks that are obtained by dividing a read image and that perform a process for correcting luminance using the value of luminance of a pixel having a largest value of luminance among pixels included in each block as a representative value of each block is not disclosed. Therefore, the correction value table is not effectively used and optimization of the quality of a corrected image is not performed.